Anthropic Simplifies Building AI Agents

Anthropic has introduced a new platform designed to streamline the creation and deployment of AI agents for enterprise clients. This offering, called Claude Managed Agents, provides developers with pre-built infrastructure to construct autonomous AI systems, effectively lowering a significant technical barrier that has hindered the automation of business workflows.

This strategic launch comes as Anthropic’s enterprise business experiences explosive growth. The company recently reported its annualized recurring revenue has exceeded $30 billion, a figure roughly triple what it was just a few months prior in December 2025. This positions Anthropic in a direct race with rivals like OpenAI, which offers its own Frontier agent platform, as both AI giants aggressively expand their robust enterprise offerings ahead of potential public listings later this year.

According to Angela Jiang, head of product for the Claude Platform, the bulk of this recent revenue surge stems from their core enterprise API product. Developers have been using it to deploy various AI agents, such as Claude Code, directly into their work environments. Jiang observes a substantial gap between the potential of Anthropic’s models and how businesses currently apply them. She states the new tool empowers any company to leverage top-tier infrastructure and deploy a team of Claude agents to handle necessary tasks.

Technically, Claude Managed Agents delivers an agent harness, a complete package of software infrastructure that enables an AI model to operate agentically by taking actions for a user. This harness integrates essential tools, a memory system, and other foundational components. Agents built with the platform also operate within a built-in sandboxed environment, allowing them to initiate software projects securely. Additional features enable developers to create agents that run autonomously in the cloud for extended periods, monitor the activity of other Claude agents, and control permissions for tool access.

Katelyn Lesse, head of engineering for the Claude Platform, highlights the engineering complexity involved. Scaling agent deployment is a complex distributed-systems engineering problem that previously required dedicated engineering teams. By providing this infrastructure ready-made, companies can redirect their engineering talent toward core business and product development instead of building underlying systems.

In a practical demonstration, the productivity startup Notion showcased its use of Managed Agents to power a client onboarding feature. Notion product manager Eric Liu illustrated how a lengthy task list within Notion could be delegated to a Claude Managed Agent, which then proceeded to complete onboarding steps sequentially. While the agent operated inside Notion, Liu monitored its progress and tool usage through a dashboard on the Claude Platform.

This expansion into enterprise tools has caught the attention of Wall Street, where some investors are growing cautious about traditional software-as-a-service companies. There is a perception that Anthropic’s broadening suite of enterprise offerings could disrupt established software vendors. Regardless of whether that threat fully materializes, the launch of Managed Agents itself underscores that widespread enterprise adoption of Claude’s ecosystem still has considerable room for growth.